Narrow Results

This work examines the use of Hybrid Intelligent Systems in the pattern recognition system of an artificial nose. The connectionist approaches Multi-Layer Perceptron and Time Delay Neural Networks, and the hybrid approaches Feature-Weighted Detector and Evolving Neural Fuzzy Networks were investigated. A Wavelet Filter is evaluated as a preprocessing method for odor signals. The signals generated by an artificial nose were composed by an array of conducting polymer sensors and exposed to two different odor databases.

A new approach for designing the Biorthogonal Wavelet Filter Bank (BWFB) for the purpose of image compression is presented in this paper. The approach is broken into two steps. First, an optimal filter bank is designed in the theoretical sense, based on Vaidyanathan's coding gain criterion in the SubBand Coding (SBC) system. Then, the above filter bank is optimized based on the criterion of Peak Signal-to-Noise Ratio (PSNR) in the JPEG2000 image compression system, resulting in a BWFB in practical application sense. With the approach, a series of BWFBs for a specific class of applications related to image compression, such as gray-level images, can be quickly designed. Here, new 7/5 BWFBs are presented based on the above approach for image compression applications. Experiments show that the 7/5 BWFBs not only have excellent compression performance, but also easy computation and are more suitable for VLSI hardware implementations. They perform equally well with respect to 7/5 filters in the JPEG2000 standard.

This paper discusses algebraic construction methods of finite orthogonal and biorthogonal wavelet filters under vanishing moment constraints, where it gives a set of algebraic constraint equations of orthogonal wavelet filter under one vanishing moment constraint, and Daubechies wavelet filters such as dbl, db2, sym2, db3 and sym3 filters are derived from these equations as examples; and it also obtains a set of constraint equations of biorthogonal wavelets under two vanishing moment constraints, (7-5) biorthogonal wavelet filter with (4,2) vanishing moment and (5-3) biorthogonal wavelet filter with (2,2) vanishing moment are derived from these equations as examples.